metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: options
sequence: string
- name: answer
dtype: string
- name: decoded_image
dtype: image
splits:
- name: train
num_bytes: 16211
num_examples: 2
download_size: 14774
dataset_size: 16211
configs:
- config_name: default
data_files:
- split: train
path: data/train-*